CPT and Pile Database for Performance-Based Design of Pile Axial Bearing Capacity-Paper

Paper

Pdf Summary

The paper presents a database-driven, reliability-based framework for selecting methods to predict axial pile bearing capacity using Cone Penetration Test (CPT) data. CPT provides continuous soil profiles and generally less measurement-related uncertainty than approaches relying on indirect estimates of soil strength. However, prediction accuracy also depends on soil variability, pile characteristics, installation and loading conditions, and the data used to develop each method.<br /><br />The authors describe the AUT:Geo-CPT&Pile database, which contains about 600 pile load-test records paired with nearby CPT or CPTu measurements. By comparing a project’s soil profile and pile details with similar database cases, designers can identify suitable prediction methods, assess uncertainty, and—in data-limited situations—select or generate relevant information. Methods are evaluated using statistical, probabilistic, and reliability criteria, including prediction bias, error, accuracy, efficiency, and wasted capacity. The resulting assessments can inform method selection and resistance factors for Load and Resistance Factor Design (LRFD).<br /><br />Examples involving piles in sand and clay illustrate the approach. Results indicate that methods such as UniCone, Meyerhof, and the Dutch method perform well in some of the examined conditions, while no method is universally best. CPT-based methods generally show stronger performance than static-analysis and SPT-based approaches in the cited comparisons.<br /><br />The paper concludes that matching database cases to site-specific conditions is essential: applying a method beyond the soil and pile conditions for which it was developed may produce misleading predictions. A systematic, database-informed selection process can reduce uncertainty, improve design efficiency, and support safer, more cost-effective, site-specific pile design.

Keywords

CPT

axial pile bearing capacity

reliability-based framework

pile load-test database

soil profile

prediction methods

LRFD

uncertainty assessment

pile design

Cone Penetration Test